An Implantable Cardioverter-Defibrillator (ICD) is a device implanted in a patient's chest to monitor the heart rhythm and deliver shocks to terminate potentially lethal arrhythmias when necessary. While ICDs reduce sudden cardiac death, patients with these devices do eventually die, either of heart failure or other chronic diseases. As a patient's disease worsens, physiologic changes (intrinsic and extrinsic to the heart) may affect the cardiac conduction system, leading to more arrhythmias and increasing the frequency of shocks. Because ICD shocks can cause psychological and physical suffering and may not prolong a life of acceptable quality, it is appropriate to consider ICD deactivation as a patient's clinical status worsens and death is near. However, these conversations rarely occur. We propose to conduct a randomized clinical trial of a physician-centered patient counseling and support intervention to improve communication between cardiologists and patients with ICDs. Most prior work on strengthening doctor-patient communication has been done in cancer, so the focus of this proposal on heart failure is novel and of high public health importance. The goals of the study are to determine the effectiveness of the intervention to: 1) increase conversations about ICD deactivation, 2) increase the number of patients who have their devices deactivated, and 3) improve mental health outcomes for bereaved caregivers of deceased patients. The unit of randomization is the hospital, the intervention is aimed at cardiologists, and the unit of analysis is the patient. We have created a network of six academic medical centers across the country. The intervention to be delivered consists of three parts. First, the PI will conduct a workshop on communication specific to ICD-deactivation with cardiologists at the intervention centers. Second, when enrolled patients are admitted to the hospital or seen in the outpatient setting, the cardiologist will receive reminders that the patient is appropriate for a conversation about ICD deactivation. Finally, cardiologists will receive quarterly aggregated feedback about the number of conversations they have conducted and data on patients' satisfaction with conversations. Physicians at usual care hospitals receive a didactic lecture on advance care planning. All patients and surrogates will be interviewed at baseline and then assessed at regular intervals to determine the prevalence of conversations about ICD deactivation and the frequency of device deactivation. Caregivers will continue to be interviewed up to 6 months after the patient dies to determine the relationship of the intervention to caregiver mental health outcomes. Given the exponential increase in the number of patients with ICDs, this intervention has the potential to improve the quality of care for thousands of patients near the end of life and their families. We also believe that using the data from this study, a similar intervention structure can be created to help physicians discuss the use of other technologic interventions in patients with advanced disease (e.g. feeding tubes for patients with advanced dementia or discontinuing left ventricular assist devices for patients with heart failure at the end of life).